Fake News Detection in Portuguese Under Large Language Model-Generated Content
DOI:
https://doi.org/10.5753/jbcs.2025.5525Keywords:
Fake news, Machine Learning, Large Language Models, Retrieval-Augmented GenerationAbstract
We are daily exposed to fake news, a growing problem that spreads in various forms, including rumours, advertisements, social media posts, and political propaganda. Predominantly created by humans, in recent years, we have witnessed an increase of digital content fabricated or manipulated with the use of deep learning. Large Language Models (LLMs), for instance, represent a real threat if used to generate highly convincing fake news that could evade conventional detection systems. This study evaluates the impact of LLM-generated fake news on machine learning (ML) classifiers. The ML models are trained with Portuguese-language datasets and experiments are conducted using aligned data, where each fake news sample has its true news counterpart. We assess the performance of each ML model with synthetic fake news, which was generated using a Portuguese-based LLM, namely Sabiá-3. Our results reveal significant performance degradation of ML models when assessed under mismatch conditions, e.g., when they are trained with human-generated content, and tested with LLM-generated fake news (or vice-versa). These findings highlight the need for updated detection strategies capable of handling the linguistic and stylistic nuances introduced by LLMs. To address that, a Retrieval-Augmented Generation (RAG) framework was evaluated under the same conditions as the ML models. The framework showed to be more robust under mismatch conditions, whereas ML models provided better performance when there was no distribution shift between train and test data. These results contribute to the understanding of fake news detection in Portuguese, emphasizing the importance of adapting existing models to the evolving nature of misleading LLM-generated content.
Downloads
References
Abonizio, H., Almeida, T. S., Laitz, T., Junior, R. M., Bonás, G. K., Nogueira, R., and Pires, R. (2024). Sabiá-3 technical report. arXiv preprint arXiv:2410.12049. Available at: [link].
Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F. L., Almeida, D., Altenschmidt, J., Altman, S., Anadkat, S., et al. (2023). Gpt-4 technical report. arXiv preprint arXiv:2303.08774. DOI: 10.48550/arxiv.2303.08774.
Aïmeur, E., Amri, S., and Brassard, G. (2023). Fake news, disinformation and misinformation in social media: a review. Social Network Analysis and Mining, 13(1):30. DOI: 10.1007/s13278-023-01028-5.
Ayoobi, N., Shahriar, S., and Mukherjee, A. (2024). Seeing through ai's lens: Enhancing human skepticism towards llm-generated fake news. In Proceedings of the 35th ACM Conference on Hypertext and Social Media, HT'24, page 1-11, New York, NY, USA. Association for Computing Machinery. DOI: 10.1145/3648188.3675136.
Azzimonti, M. and Fernandes, M. (2023). Social media networks, fake news, and polarization. European Journal of Political Economy, 76:102256. DOI: 10.1016/j.ejpoleco.2022.102256.
Bojanowski, P., Grave, E., Joulin, A., and Mikolov, T. (2017). Enriching word vectors with subword information. Transactions of the Association for Computational Linguistics, 5:135-146. DOI: 10.1162/tacl_a_00051.
Boser, B. E., Guyon, I. M., and Vapnik, V. N. (1992). A training algorithm for optimal margin classifiers. pages 144-152. DOI: 10.1145/130385.130401.
Breiman, L. (2001). Random forests. ml, 45(1):5-32. DOI: 10.1023/A:1010933404324.
Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J. (1984). Classification and Regression Trees. Wadsworth International Group, Belmont, California, USA. Book.
Cabral, L., Monteiro, J. M., da Silva, F., Wellington, J., Mattos, C. L. C., and Mourão, P. J. C. (2021). Fakewhastapp.br: Nlp and machine learning techniques for misinformation detection in brazilian portuguese whatsapp messages. In Proceedings of the 23rd International Conference on Enterprise Information Systems (ICEIS 2021) - Volume 1, pages 63-74. DOI: 10.5220/0010446800630074.
Chavarro, J., Carvalho, J., Portela, T., and Silva, J. (2023). Faketruebr: Um corpus brasileiro de notícias falsas. In Anais da XVIII Escola Regional de Banco de Dados, pages 108-117, Porto Alegre, RS, Brasil. SBC. DOI: 10.5753/erbd.2023.229495.
Cordeiro, P. R. and Pinheiro, V. (2019). Um corpus de notícias falsas do twitter e verificaçao automática de rumores em língua portuguesa. In Proceedings of the Symposium in Information and Human Language Technology, pages 219-228. DOI: 10.5753/stil.2021.17796.
Cortes, C. and Vapnik, V. N. (1995). Support-vector networks. ml, 20(3):273-297. DOI: 10.1007/BF00994018.
DeepSeek-AI (2025). Deepseek-r1: incentivizing reasoning capability in llms via reinforcement learning. arXiv preprint arXiv:2501.12948. Available at: [link].
Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). Bert: Pre-training of deep bidirectional transformers for language understanding. In Burstein, J., Doran, C., and Solorio, T., editors, Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4171-4186, Minneapolis, Minnesota. Association for Computational Linguistics. DOI: 10.18653/v1/N19-1423.
Douze, M., Guzhva, A., Deng, C., Johnson, J., Szilvasy, G., Mazaré, P.-E., Lomeli, M., Hosseini, L., and Jégou, H. (2025). The faiss library. arXiv preprint arXiv:2401.08281. DOI: 10.1109/tbdata.2025.3618474.
Ecker, U. K., Lewandowsky, S., Cook, J., Schmid, P., Fazio, L. K., Brashier, N., Kendeou, P., Vraga, E. K., and Amazeen, M. A. (2022). The psychological drivers of misinformation belief and its resistance to correction. Nature Reviews Psychology, 1(1):13-29. DOI: 10.1038/s44159-021-00006-y.
Faustini, P. and Covões, T. F. (2019). Fake news detection using one-class classification. In Proceedings of the 8th Brazilian Conference on Intelligent Systems (BRACIS'19), pages 592-597, Salvador, BA, Brazil. IEEE. DOI: 10.1109/BRACIS.2019.00109.
Gallotti, R., Valle, F., Castaldo, N., Sacco, P., and De Domenico, M. (2020). Assessing the risks of 'infodemics' in response to covid-19 epidemics. Nature human behaviour, 4(12):1285-1293. DOI: 10.1038/s41562-020-00994-6.
Garcia, G. L., Afonso, L. C., and Papa, J. P. (2022). Fakerecogna: a new brazilian corpus for fake news detection. In International Conference on Computational Processing of the Portuguese Language, pages 57-67. Springer. DOI: 10.1007/978-3-030-98305-5_6.
Garcia, G. L., Paiola, P. H., Jodas, D. S., Sugi, L. A., and Papa, J. P. (2024). Text summarization and temporal learning models applied to portuguese fake news detection in a novel brazilian corpus dataset. In Gamallo, P., Claro, D., Teixeira, A., Real, L., Garcia, M., Oliveira, H. G., and Amaro, R., editors, Proceedings of the 16th International Conference on Computational Processing of Portuguese - Vol. 1, pages 86-96, Santiago de Compostela, Galicia/Spain. Association for Computational Lingustics. Available at: [link].
Geurgas, R. and Tessler, L. R. (2024). Automatic detection of fake tweets about the covid-19 vaccine in portuguese. Social Network Analysis and Mining, 14(1):55. DOI: 10.1007/s13278-024-01216-x.
Gôlo, M. P. S., Mori, A. L. V., Oliveira, W. G., Barbosa, J. R., Graciano-Neto, V. V., de Lima, E. A., and Marcacini, R. M. (2024). On the use of large language models to detect brazilian politics fake news. In Proceedings of the 21st National Meeting on Artificial and Computational Intelligence (ENIAC'2024), pages 1-12, Belém, PA, Brazil. Brazilian Computer Society. DOI: 10.5753/eniac.2024.245119.
Jegham, N., Abdelatti, M., Elmoubarki, L., and Hendawi, A. (2025). How hungry is ai? benchmarking energy, water, and carbon footprint of llm inference. arXiv preprint arXiv:2505.09598. Available at: [link].
Jiang, B., Tan, Z., Nirmal, A., and Liu, H. (2024). Disinformation detection: An evolving challenge in the age of llms. In Proceedings of the 2024 SIAM International Conference on Data Mining (SDM), pages 427-435. SIAM. DOI: 10.1137/1.9781611978032.50.
Kaliyar, R. K., Goswami, A., and Narang, P. (2021). Fakebert: Fake news detection in social media with a bert-based deep learning approach. Multimedia tools and applications, 80(8):11765-11788. DOI: 10.1007/s11042-020-10183-2.
Lazer, D. M., Baum, M. A., Benkler, Y., Berinsky, A. J., Greenhill, K. M., Menczer, F., Metzger, M. J., Nyhan, B., Pennycook, G., Rothschild, D., et al. (2018). The science of fake news. Science, 359(6380):1094-1096. DOI: 10.1126/science.aao2998.
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-t., Rocktäschel, T., Riedel, S., and Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive nlp tasks. In Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS '20, Red Hook, NY, USA. Curran Associates Inc.. DOI: 10.48550/arxiv.2005.11401.
Mikolov, T., Sutskever, I., Chen, K., Corrado, G., and Dean, J. (2013). Distributed representations of words and phrases and their compositionality. In Proceedings of the 26th International Conference on Neural Information Processing Systems (NIPS'13), pages 3111-3119, Lake Tahoe, Nevada, USA. Curran Associates Inc.. DOI: 10.48550/arXiv.1310.4546.
Monteiro, R. A., Santos, R. L. S., Pardo, T. A. S., de Almeida, T. A., Ruiz, E. E. S., and Vale, O. A. (2018). Contributions to the study of fake news in portuguese: New corpus and automatic detection results. In 13th International Conference on Computational Processing of the Portuguese Language (PROPOR'2018), pages 324-334, Canela, Rio Grande do Sul, Brazil. Springer International Publishing. DOI: 10.1007/978-3-319-99722-3_33.
O'Connor, C. and Weatherall, J. O. (2021). Modeling how false beliefs spread. The Routledge Handbook of Political Epistemology, pages 203-213. DOI: 10.4324/9780429326769-25.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikit-learn: Machine learning in python. Journal of Machine Learning Research, 12:2825-2830. DOI: 10.5555/1953048.2078195.
Pires, R., Abonizio, H., Almeida, T. S., and Nogueira, R. (2023). Sabiá: Portuguese large language models. In Naldi, M. C. and Bianchi, R. A. C., editors, Intelligent Systems, pages 226-240, Cham. Springer Nature Switzerland. DOI: 10.48550/arXiv.2304.07880.
Plikynas, D., RizgelienÄ—, I., and Korvel, G. (2025). Systematic review of fake news, propaganda, and disinformation: Examining authors, content, and social impact through machine learning. IEEE Access, 13:17583-17629. DOI: 10.1109/ACCESS.2025.3530688.
Rathore, F. A. and Farooq, F. (2020). Information overload and infodemic in the covid-19 pandemic. J Pak Med Assoc, 70(5):S162-S165. DOI: 10.5455/jpma.38.
Reimers, N. and Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics. DOI: 10.18653/v1/d19-1410.
Sardinha, T. B. (2024). Ai-generated vs human-authored texts: A multidimensional comparison. Applied Corpus Linguistics, 4(1):100083. DOI: 10.1016/j.acorp.2023.100083.
Saroj, A. and Pal, S. (2020). Use of social media in crisis management: A survey. International Journal of Disaster Risk Reduction, 48:101584. DOI: 10.1016/j.ijdrr.2020.101584.
Shu, K., Sliva, A., Wang, S., Tang, J., and Liu, H. (2017). Fake news detection on social media: A data mining perspective. ACM SIGKDD explorations newsletter, 19(1):22-36. DOI: 10.48550/arXiv.1708.01967.
Silva, R. M., de Sales Santos, R. L., Pardo, T. A. S., and Almeida, T. A. (2020). Towards automatically filtering fake news in portuguese. Expert Systems with Applications, 146:1-48. DOI: 10.1016/j.eswa.2020.113199.
Silva, R. M., Pires, P. R., and Almeida, T. A. (2023). Incremental learning for fake news detection. Journal of Information and Data Management, 13(6). DOI: 10.5753/jidm.2022.2542.
Sivakumar, S., Videla, L. S., Kumar, T. R., Nagaraj, J., Itnal, S., and Haritha, D. (2020). Review on word2vec word embedding neural net. In 2020 international conference on smart electronics and communication (ICOSEC), pages 282-290. IEEE. DOI: 10.1109/icosec49089.2020.9215319.
Smith, S. T., Kao, E. K., Mackin, E. D., Shah, D. C., Simek, O., and Rubin, D. B. (2021). Automatic detection of influential actors in disinformation networks. Proceedings of the National Academy of Sciences, 118(4):e2011216118. DOI: 10.1073/pnas.2011216118.
Souza, F., Nogueira, R., and Lotufo, R. (2020). Bertimbau: pretrained bert models for brazilian portuguese. In Cerri, R. and Prati, R. C., editors, Proceedings of the 9th Brazilian Conference on Intelligent Systems, BRACIS, Rio Grande do Sul, Brazil, October 20-23, pages 403-417, Cham. Springer International Publishing. DOI: 10.1007/978-3-030-61377-8_28.
Su, J., Zhuo, T. Y., Mansurov, J., Wang, D., and Nakov, P. (2023). Fake news detectors are biased against texts generated by large language models. arXiv preprint. DOI: 10.48550/arXiv.2309.08674.
Tomlinson, B., Black, R. W., Patterson, D. J., and Torrance, A. W. (2024). The carbon emissions of writing and illustrating are lower for ai than for humans. Scientific Reports, 14. DOI: 10.1038/s41598-024-54271-x.
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., Rodriguez, A., Joulin, A., Grave, E., and Lample, G. (2023). Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971. Available at: [link].
Vosoughi, S., Roy, D., and Aral, S. (2018). The spread of true and false news online. Science, 359(6380):1146-1151. DOI: 10.1126/science.aap9559.
Yu, H.-F., Huang, F.-L., and Lin, C.-J. (2011). Dual coordinate descent methods for logistic regression and maximum entropy models. ml, 85(1-2):41-75. DOI: 10.1007/s10994-010-5221-8.
Zhang, T., Wu, F., Katiyar, A., Weinberger, K. Q., and Artzi, Y. (2020). Revisiting few-sample bert fine-tuning. CoRR, abs/2006.05987. Available at: [link].
Zhou, L., Burgoon, J., Twitchell, D., Qin, T., and Nunamaker Jr., J. (2004). A comparison of classification methods for predicting deception in computer-mediated communication. Journal of Management Information Systems, 20(4):139-165. DOI: 10.1080/07421222.2004.11045779.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Renato Moraes Silva, Hazem Amamou, Lucca Baptista Silva Ferraz, Fabio Kauê Araujo da Silva, Anderson Raymundo Avila
This work is licensed under a Creative Commons Attribution 4.0 International License.
